Rice University published a joint paper with Microsoft on image sensor power consumption, to be presented at ACM Int. Conf. Mobile Systems, Applications and Services (MobiSys) in Taipei, Taiwan on June 25-28, 2013:
"Energy characterization and optimization of image sensing toward continuous mobile vision"
Robert LiKamWa, Bodhi Priyantha, Matthai Philipose, Lin Zhong, and Paramvir Bahl
The power consumption of five image sensors from two major vendors in the mobile market is analysed, breaking down the power consumption by major components and by operational modes. Few interesting conclusions:
"Energy characterization and optimization of image sensing toward continuous mobile vision"
Robert LiKamWa, Bodhi Priyantha, Matthai Philipose, Lin Zhong, and Paramvir Bahl
The power consumption of five image sensors from two major vendors in the mobile market is analysed, breaking down the power consumption by major components and by operational modes. Few interesting conclusions:
- Modern image sensors are not energy-proportional: energy per pixel is in fact inversely proportional to frame rate and resolution of image capture, and thus image sensor systems fail to provide an important principle of energy-aware system design: trading quality for energy efficiency.
- Optimal clock scaling reduces power by up to 50% or 30% for one mega-pixel photos and videos, respectively;
- By entering low-power standby mode between frames, an image sensor achieves almost constant energy per pixel for video capture at low frame rates, results in 40% power reduction for 1MP, 5fps capture.
